0
$\begingroup$

So I'm given a matrix $A$

$$A=\begin{pmatrix}0.4 & 0.3& 0.3\\ 0.3 & 0.5 & 0.2 \\ 0.3& 0.2 & 0.5 \end{pmatrix}$$

Find the eigenvector corresponding to the eigenvalue $\lambda_1=1$ then proceed to find the remaining eigenvalues using a built in command in MatLab

So far I've found the eigenvector for $\lambda_1=1$ which is $v_1=\begin{pmatrix}1\\1\\1\end{pmatrix}$ and after i've used the eig command I found the remaining eigenvalues to be $\lambda_2=0.3,\lambda_3=0.1$ now it wants me to literary "guess" the corresponding eigenvectors.And this is where i'm stuck. I have read lots of guides and stuff about how to find eigenvectors by inspection but that was for 2x2 matrices and for 3x3 matrices they just mentioned that they need a 3rd fact which is a bit more complicated. Besides I can't see any obvious eigenvectors.....

$\endgroup$
  • 1
    $\begingroup$ Where does it say "guess"? eig will give you both eigenvalues and eigenvectors. $\endgroup$ – Robert Israel Apr 8 at 19:50
  • $\begingroup$ @RobertIsrael In my problem sheet. (Use "eig" in matlab to find the remaining eigenvalues). From what the command returns, guess the corresponding eigenvectors then verify them (the verifying part is not an issue) but the part that it says "guess" got me pretty confused $\endgroup$ – Sami Shafi Apr 8 at 19:55
  • $\begingroup$ hmm maybe I've misunderstood the problem? As you said you'll get both eigenvectors and values if you use [v,b] = eig maybe it's referring to guess which one of the eigenvalues is for which eigenvectors as in you get both of them so you just somehow "see" which of them is for which? $\endgroup$ – Sami Shafi Apr 8 at 20:00
1
$\begingroup$

Not really "guesses", but...

Note that $$A - 0.3 I = \pmatrix{0.1 & 0.3 & 0.3\cr 0.3 & 0.2 & 0.2\cr 0.3 & 0.2 & 0.2\cr} $$ has its second and third columns equal. That says that $\pmatrix{0\cr 1 \cr -1\cr}$ will be in its null space, i.e. an eigenvector of $A$ for eigenvalue $0.3$.

Now $A$ being a symmetric matrix, its eigenvectors are orthogonal, so the eigenvector for the remaining eigenvalue $0.1$ has sum $0$ and its second and third entries equal. Thus it should be (a multiple of) $\pmatrix{2 \cr -1\cr -1\cr}$.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.